Predicting cancer outcomes from histology and genomics using convolutional networks.

In the GSCNN multivariable regression model, molecular subtype was not significant, but histologic grade was marginally significant.

also in 26,082 other papers

In the SCNN regression analysis, histologic-grade variables were not significant, but molecular subtype variables were highly significant, indicating that SCNN could capture histologic information from image data but could not learn molecular subtype information entirely from histology.

also in 132,142 other papers